Академический Документы
Профессиональный Документы
Культура Документы
ITU-T
T e l e c o m m u n i c a t i o n
U n i o n
G.8271/Y.1366
TELECOMMUNICATION
STANDARDIZATION SECTOR
OF ITU
Amendment 2
(01/2015)
G.100G.199
G.200G.299
G.300G.399
G.400G.449
G.450G.499
G.600G.699
G.700G.799
G.800G.899
G.900G.999
G.1000G.1999
G.6000G.6999
G.7000G.7999
G.8000G.8999
G.8000G.8099
G.8100G.8199
G.8200G.8299
G.8600G.8699
G.9000G.9999
Summary
Amendment 2 to Recommendation ITU-T G.8271/Y.1366 (2012) provides the following update:
History
Approval
Study Group
Unique ID*
2012-02-13
15
1.1
15
1.2
15
11.1002/1000/11527
11.1002/1000/12033
11.1002/1000/12391
Edition
1.0
Recommendation
ITU-T G.8271/Y.1366
____________________
*
To access the Recommendation, type the URL http://handle.itu.int/ in the address field of your web
browser, followed by the Recommendation's unique ID. For example, http://handle.itu.int/11.1002/1000/11
830-en.
Rec. ITU-T G.8271/Y.1366 (2012)/Amd.2 (01/2015)
FOREWORD
The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of
telecommunications, information and communication technologies (ICTs). The ITU Telecommunication
Standardization Sector (ITU-T) is a permanent organ of ITU. ITU-T is responsible for studying technical,
operating and tariff questions and issuing Recommendations on them with a view to standardizing
telecommunications on a worldwide basis.
The World Telecommunication Standardization Assembly (WTSA), which meets every four years, establishes
the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these topics.
The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1.
In some areas of information technology which fall within ITU-T's purview, the necessary standards are
prepared on a collaborative basis with ISO and IEC.
NOTE
In this Recommendation, the expression "Administration" is used for conciseness to indicate both a
telecommunication administration and a recognized operating agency.
Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain
mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the
Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other
obligatory language such as "must" and the negative equivalents are used to express requirements. The use of
such words does not suggest that compliance with the Recommendation is required of any party.
ITU 2015
All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior
written permission of ITU.
ii
Figure I.1 illustrates the delays between a packet slave clock function, or requestor (denoted as slave
throughout this clause), and a packet master clock function, or responder (denoted as master
throughout this clause). The mean propagation delay is measured at the slave after exchange of event
messages. If the Delay Request and the Delay Response mechanism is used (see [IEEE 1588-2008]),
the slave sends Delay_Req and the master sends Delay_Resp and, separately, Sync and Follow_Up
(i.e., the sending of Sync and Follow_Up are not part of the Delay_Req/Delay_Resp exchange; the
Follow_Up message is sent if, and only if, the clock is two-step). If the Peer Delay mechanism is used
(see [IEEE 1588-2008]), the slave sends Pdelay_Req and the master sends Pdelay_Resp and, if the
clock is two-step, Pdelay_Resp_Follow_Up.
The figure shows the effective points in the protocol stack of each clock where timestamps are
generated, after any corrections for ingress and egress latencies are made (see section 7.3.4 and
Figure 19 of [IEEE 1588-2008]). These points would ideally be at the reference plane, i.e., the
boundary point between the PHY and the network physical medium. However, in practice, the
corrections for ingress and egress latencies are not perfect, and the effective points at which the
timestamps are generated differ from the reference plane. The delays between the effective points
where timestamps are taken and the reference plane are denoted dtxPHY,M and drxPHY,M for egress and
ingress, respectively, at the master, and dtxPHY,S and drxPHY,S for egress and ingress, respectively, at the
slave. In this notation, the subscript t (transmit) is used for egress and the subscript r (receive) is used
for ingress. In general, these four quantities can all be different.
The figure also shows the link delays, which are measured from the reference plane of one clock to
the reference plane of the other clock. The delay from the master to the slave is denoted dmslink, and
the delay from the slave to the master is denoted dsmlink.
Figure I.1 Illustration of delays between a packet slave clock function, or requestor,
and a packet master clock function, or responder
The total delay from the master to the slave, tms, is the sum of the delays in that direction:
link
PHY, S
t ms d txPHY,M d ms
d rx
(I-1)
Similarly, the total delay from the slave to the master, tsm, is the sum of the delays in that direction:
link
PHY, M
t sm d txPHY,S d sm
d rx
(I-2)
For the sign convention for the delay asymmetry, the same convention as in section 7.4.2 of
[IEEE 1588-2008] is adopted. Let Dmean denote the measured mean path delay (i.e., the measured
result of the exchange of Delay_Req and Delay_Resp or of Peer Delay messages), and Dasym denote
the total delay asymmetry. Then, Dasym is defined to be positive when the delay from the master to
the slave is larger than the delay from the slave to the master. Likewise, Dasym is defined to be negative
when the delay from the master to the slave is smaller than the delay from the slave to the master.
Then:
t ms Dmean Dasym
t sm Dmean Dasym
(I-3)
Dmean
t ms t sm
2
(I-4)
as required. Substituting equations (I-1) and (I-2) into equation (I-4) gives:
Dmean
link
PHY, S
link
PHY, M
(d txPHY, M d ms
d rx
) (d txPHY, S d sm
d rx
)
2
(I-5)
Either of the two equations (I-3) may be used with equation (I-4) to obtain the delay asymmetry in
terms of the component delays. Using the first of equations (I-3) produces:
Dasym t ms Dmean
(d txPHY, M
link
d ms
PHY, S
d rx
)
link
PHY, S
link
PHY, M
d rx
d rx
(d txPHY, M d ms
) (d txPHY, S d sm
)
PHY, M
link
link
PHY, S
d sm
d txPHY, S
d txPHY, M d rx
d ms
d rx
2
2
2
S
eM
phy elink asym e phy
(I-6)
where:
PHY, M
d txPHY, M d rx
2
(I-7)
link
link
d ms
d sm
elinkasym
2
(I-8)
PHY, S
d txPHY,S d rx
2
(I-9)
eM
phy
e Sphy
Equations (I-7) and (I-9) are the errors due to PHY latency asymmetry at the master and slave
respectively. Equation (I-8) is the error due to link asymmetry. Equation (I-6) indicates that, in
computing the total asymmetry, the errors due to PHY latency at the master and due to the link are
added, while the error due to PHY latency at the slave is subtracted.
2)
Appendix IV
Link and network asymmetry compensation
(This appendix does not form an integral part of this Recommendation.)
In order to compensate for link delay asymmetry, it might be desirable to have in place some
automatic link asymmetry calibration procedure. This could be based on calculating the propagation
delays by means of two-way measurements made on the fibres used by the traffic.
The procedure can be done separately on both fibres (in the fibre used in the forward direction and in
the fibre used for the reverse direction) providing the forward propagation delay df and the reverse
propagation delay dr. This is shown in Figure IV.1.
Figure IV.1 Link asymmetry calibration process (performed separately on both fibres)
Alternatively the round trip measurement could be done in two steps on both fibres by reversing the
direction of transmission. This is shown in Figure IV.2.
df = dms
dr = dsm
The link asymmetry calibration mechanism must meet an accuracy objective for df and dr estimations.
This limit is for further study.
NOTE 2 In the case during the asymmetry calculation procedure where one node enters holdover (e.g.,
caused by the fibres-swapping if this is required by the procedure), the effect of the frequency holdover needs
to be taken into account as it might impact the accuracy of the measurement.
Several implementations are possible, e.g., based on optical switches or fixed or tunable add drop
filters. Depending on the implementation, it may not be required to interrupt the traffic during the
calibration process and hence in-service operation might be possible. However, the asymmetry
compensation is a process that is only required at start-up or during rearrangements in the network.
This measurement is applicable for WDM systems (including OTN) and non-WDM systems. In the
case of wavelength-division-multiplexing (WDM) systems, this measurement should also take into
account possible delay due to dispersion-compensating fibre (DCF).
NOTE 3 In the case of WDM systems, the asymmetry due to the use of different wavelengths in the two
directions should also be taken into account. Indeed, the use of different wavelengths on the two fibres, (or in
a single fibre in the case of a transmission system using a single fibre), would result in different delays even if
the fibres have the same length. Note also that a compensation related to the same aspect would be required if
the wavelength used during the link asymmetry calibration process is different from the wavelength used by
the traffic. Suitable methodologies to address this point are introduced in Appendix III.
The difference (df dr) can be used in the evaluation of the delay asymmetry to be used in the time recovering
process. In particular the delayAsymmetry parameter as defined in section 7.4 of [IEEE 1588-2008] would be
half of that difference.
NOTE 4 If a T-BC is implemented in every node, the compensation can be triggered directly by the T-BC,
which would know the difference (df dr). If this is not the case, some means have to be provided in order to
make the difference (df dr) available at the points in the network where the precision time protocol (PTP)
messages are processed. This is for further study.
NOTE 5 In the case of a time synchronization carried by PTP, the PTP connection may have asymmetry due
to a variety of reasons, including network paths, loading levels or cable lengths. The asymmetry of a PTP
connection may be evaluated at a PTP network element, if the network element has access to a second time
synchronization source that is not significantly impacted by asymmetry (such as a GNSS receiver, or a time
synchronization reference carried via timing protocols such as PTP with proper accuracy) as shown in
Figure IV.3. If the asymmetry of the PTP connection is evaluated using such a second time synchronization
source, then the offset caused by the asymmetry may be compensated by the network element. The same
principle could be applied between network elements in a chain.
Y.100Y.199
Y.200Y.299
Y.300Y.399
Y.400Y.499
Y.500Y.599
Y.600Y.699
Y.700Y.799
Y.800Y.899
Y.1000Y.1099
Y.1100Y.1199
Y.1200Y.1299
Y.1300Y.1399
Y.1400Y.1499
Y.1500Y.1599
Y.1600Y.1699
Y.1700Y.1799
Y.1800Y.1899
Y.1900Y.1999
Y.2000Y.2099
Y.2100Y.2199
Y.2200Y.2249
Y.2250Y.2299
Y.2300Y.2399
Y.2400Y.2499
Y.2500Y.2599
Y.2600Y.2699
Y.2700Y.2799
Y.2800Y.2899
Y.2900Y.2999
Y.3000Y.3499
Y.3500Y.3999
Series D
Series E
Overall network operation, telephone service, service operation and human factors
Series F
Series G
Series H
Series I
Series J
Cable networks and transmission of television, sound programme and other multimedia
signals
Series K
Series L
Construction, installation and protection of cables and other elements of outside plant
Series M
Series N
Series O
Series P
Series Q
Series R
Telegraph transmission
Series S
Series T
Series U
Telegraph switching
Series V
Series X
Series Y
Series Z
Printed in Switzerland
Geneva, 2015